moments of correlated gamma variates

10
 This article was downloaded by: [The University of Manchester Library] On: 22 October 2013, At: 13:58 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Communications in Statistics - Theory and Methods Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/lsta20 Moments for a ratio of correlated gamma variates J.D. Tubbs a a  Department of Mathematical Sciences , University of Arkansas Published online: 27 Jun 2007. To cite this article:  J.D. T ubbs (1986) Moments for a ratio of correlated gamma variates, Communications in Statistics - Theory and Methods, 15:1, 251-259 To link to this article: http://dx.doi.org/10.1080/03610928608829119 PLEASE SCROLL DOWN FOR ARTICLE T aylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representation s or warranties whatsoever as to the accuracy, completene ss, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by T aylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. T aylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistributi on, reselling, loan, sub-licensing, systemat ic supply, or distribution in any form to anyone is expressly forbidden. T erms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Upload: azwan-mahmud

Post on 07-Oct-2015

225 views

Category:

Documents


0 download

DESCRIPTION

Moments of Correlated Gamma variates

TRANSCRIPT

  • This article was downloaded by: [The University of Manchester Library]On: 22 October 2013, At: 13:58Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

    Communications in Statistics - Theory and MethodsPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/lsta20

    Moments for a ratio of correlated gamma variatesJ.D. Tubbs aa Department of Mathematical Sciences , University of ArkansasPublished online: 27 Jun 2007.

    To cite this article: J.D. Tubbs (1986) Moments for a ratio of correlated gamma variates, Communications in Statistics -Theory and Methods, 15:1, 251-259

    To link to this article: http://dx.doi.org/10.1080/03610928608829119

    PLEASE SCROLL DOWN FOR ARTICLE

    Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shall not beliable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out ofthe use of the Content.

    This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

  • COMMUN. STATIST.-THEORY METH., 1 5 ( 1 ) , 251-259 ( 1986 )

    MOMENTS FOR A RATIO OF CORRELATED GA'lMA VARIATES

    J . D . Tubbs

    Department of K a t h e m a t i c a l S c i e n c e s U n i v e r s i t y of Arkansas

    Key Words and P h r a s e s : h y p e r g e o m z t ~ ~ e fz,nctions; Kwmer's i d e n t i t y ; bivar

  • TUBBS

    In this paper the results are restricted to the moments of

    positively correlated gamma distributed variates. Flueck, Holland,

    and Lee (1975, 1979) presented some exact distributional results for the ratio, R, of correlated gamma under the Cherian-David-Fix

    bivariate gamma distributional structure. Recently, Tubbs and

    Smith (1984) obtained comparable results for R using a special case of Jensen's bivariate gamma distributional structure. The

    corresponding expressions for the moments of R from the two struc- tures were quite different in their appearance.

    This paper investigates the comparison of the respective

    finite moments whenever they exist. This comparison is performed

    both analytically and numerically. In section 2, the results for each of the specified bivariate gamma structures is reviewed.

    Section 3 contains the analytical results for the comparison of

    the expressions for the respective moments. The numerical results

    are presented in section 4.

    2. DISTRZBUTTON AXD NPIENTS OF R - - - - - -.- - . - - - - - -

    2.1 Cherian-David-Xix-S tructure

    The results in this section are found in Lee, Holland, and

    Flueck (1979) and llielke and Flueck (1976). Using the notation of ?Iielke and Flueck, the probability density function for R = X / Y

    where X and Y are positively correlated gamma variates is given by

    (a)m+n(b)m(c)n where F (a,b,c,d:x,y) = C ----------- 1 (d),+,m!n! X Y ,/xl

  • MOMENTS FOR RATIO OF CORRELATED GAMMA VARIATES 253

    X = U+P, Y = V+P. U, V, and P a r e independent gamma random v a r i a -

    b l e s with common s c a l e parameter X and r e s p e c t i v e shape parameters

    a-5, B - S , and S(O

  • 254 TUBBS

    The purpose of t h i s p a p e r i s t o c o n s i d e r b o t h e x p r e s s i o n s

    g i v e n i n e q u a t i o n s ( 2 . 2 ) and (2 .5) and d e t e r m i n e unde r which c o n d i t i o n s t h e two e q n a t i o n s p r o v i d e i d e n t i c a l r e s u l t s . Consid-

    e r i n g e q u a t i o n (2 .5 ) and t h e d e f i n i t i o n and p r o p e r t i e s of one and two d i m e n s i o n a l hype rgeomet r i c f u n c t i o n s a l o n g w i t h Kummer's

    i d e n t i t y , e q u a t i o n ( 2 . 5 ) can b e r e w r i t t e n a s e i t h e r

    Tn o r d e r t o perform a compar ison of t h e e x p r e s s i o n f o r t h e

    moments g i v e n b y t h e two mr thods , e q u a t i o n ( 2 . 2 ) w i l l be r e w r i t t e n a s

    f o r some c o e f f i c i e n t s 5 . Fur the rmore , e q u a t i o n (2.6) can be j w r i t t e n a s

    where

    Using u ( 5 ) and vrn(C), i t f o l l o w s t h a t t h e two methods a r e iden - m

    t i c a l i f a n d o n l y i f a = b . f o r a l l j ' s . Fur the rmore , s i n c e j J

    Dow

    nloa

    ded

    by [T

    he U

    nivers

    ity of

    Man

    ches

    ter L

    ibrary

    ] at 1

    3:58 2

    2 Octo

    ber 2

    013

  • MOMENTS FOR RATIO OF CORRELATED GAMMA VARIATES 255

    i t fo l lows t h a t (2.2) and (2.6) a r e equa l i f and only i f h m ( j ) = k ( j ) f o r a l l j = O , l , . . . ,m < B . The above n o t a t i o n w i l l b e used

    m

    i n proving t h e main r e s u l t s a s summarized i n t h e fol lowing theorem.

    Theorem: Expressions f o r t h e mth unad jus ted popula t ion moments --

    f o r R , t h e r a t i o of c o r r e l a t e d gamma v a r i a t e s given by equa t ions

    (2.2) and (2.6) a r e i d e n t i c a l whenever e i t h e r i ) X and Y a r e independent and m E (0,R) o r i i ) m = 1 f o r B > 1.

    Proof: i) I f X and Y a r e independent then 5 = 0, which i m p l i e s --

    t h e equa t ions (3.1) and (3.2) reduce t o km(0) and hm(0) , respec- t i v e l y , f o r any nonnegat ive i n t e g e r m(m< a ) . From equat Ion (3.31, i t fo l lows t h a t h,(O) = ( U ) ~ ( R ) - ~ . By cons ider ing u m ( 5 ) i n equa- t i o n (3.1) and n o t i n g t h a t (a-5).(5),j = 0 i f 5 = 0 and J j = 0 , 1 , 2 , . . .m-1 and t h a t (a-5) . ( s ) , ~ = f o r j = m and

    J 5 = 0, hence

    I f m = 1 and 5 > 0 then equa t ions (2.2) and (2.6) a r e iden- t i c a l i f and only if h l ( j ) = k l ( j ) f o r j = 0 , l . It fo l lows from (3.3) t h a t

    1 h l ( 0 ) = ~ I R - 1 , h l ( l ) = B(D-~).

    From (3.5) it fol lows t h a t

    Dow

    nloa

    ded

    by [T

    he U

    nivers

    ity of

    Man

    ches

    ter L

    ibrary

    ] at 1

    3:58 2

    2 Octo

    ber 2

    013

  • 2 5 6

    Since ( x ) ~ = 1 and ( x ) ~ = x, i t i s e a s i l y shown t h a t

    TUBBS

    i i ) Suppose t h a t 1 < m < B and t h a t 5 > 0. The two expres- s i o n s a r e i d e n t i c a l whenever k m ( j ) = hm(j ) f o r j = 1 , 2 , . . . , m . Consider t h e c a s e where j = m. Again from (3 .3) i t fol lows t h a t

    I n cons ider ing km(m) from equa t ion (3 .5 ) , i t is s u f f i c i e n t t o con- m

    s i d e r t h e c o e f f i c i e n t 5 i n t h e expansion (a-5) . (5 ) f o r any J m-j

    f i x e d j . Since t h i s c o e f f i c i e n t i s always ( - I ) ' , it fo l lows t h a t

    In cons ider ing equa t ions (3.7) and ( 3 . 8 ) , one observes t h a t k (m) i s a f u n c t i o n of t h e parameter a v~hereas hm(m) i s indepen-

    m

    dent of a . Hence, equa t ions ( 2 . 2 ) and (2.6) a r e n o t i d e n t i c a l f o r a r b i t r a r y parameters a and 6.

    The above theorem g i v e s necessary and s u f f i c i e n t condi t ions f o r t h e two methods t o provide i d e n t i c a l r e s u l t s . The next s e c t i o n cons iders a comparison of t h e two methods whenever t h e

    methods d i f f e r . The comparison i s made us ing a numerical eval- u a t i o n of t h e two methods f o r a s e l e c t e d s u b s e t of t h e parameter space.

    4 . NUMERICAL EVALUATE

    In t h i s s e c t i o n , equa t ions (2.2) and (2.6) a r e eva lua ted numer ica l ly u s i n g a s e l e c t e d s u b s e t of t h e parameter space. D

    ownl

    oade

    d by

    [The

    Univ

    ersity

    of M

    anch

    ester

    Libr

    ary] a

    t 13:5

    8 22 O

    ctobe

    r 201

    3

  • MOMENTS FOR RATIO OF CORRELATED GAMMA VARIATES

    METHOD 1 METHOD 2

    4 0 . 2 5 0 . 16 l-I . 11 4 0 . 0 5

    1 . @ 0 (1.67 0 . 3 6 0 . 2 0

    8 . 7 5 5 . 4 1 3 . 1 9 1.79

    I . 3 l 0.96 0 . 6 3 0 . 4 9

    0 . 4 2 0.33 0 . 2 6 0 .20

    t o . 10 0 . OH 0. 07 0.116

    2 . t i2 2 . (:I 8 1.64 1.28

    o .60 0 .52 0 .44 O. 3 8

    I .87 1.60 1.37 1 . 1 7

    Dow

    nloa

    ded

    by [T

    he U

    nivers

    ity of

    Man

    ches

    ter L

    ibrary

    ] at 1

    3:58 2

    2 Octo

    ber 2

    013

  • 258 TUBBS

    Tab le 1 summarizes t h e f i r s t f o u r u n a d j u s t e d p o p u l a t i o n moments. Method 1 r e f e r s t o e q u a t i o n ( 2 . ~ ) and Method 2 r e f e r s t o equa- t i o n ( 2 . 6 ) .

    From Tab le 1, one o b s e r v e s t h a t t h e h i g h e r unad jus ted moments a r e d i f f e r e n t f o r t h e two methods whenever t h e v a r i a b l e s X and Y a r e c o r r e l a t e d . However, t h e r e is v e r y l i t t l e d i f f e r e n c e i n t h e f i r s t two moments. From t h e t a b l e one a l s o o b s e r v e s t h a t t h e u n a d j u s t e d moments a r e s i g n i f i c a n t l y a f f e c t e d by t h e p r e s e n c e of c o r r e l a t i o n . Th i s phenomenon was obse rved by Flueck and Hol land (1976) i n e s t i m a t i n g t h e moments of r a t i o s u s i n g r a i n - f a l l d a t a . Although t h e e x p r e s s i o n s f o r t h e s e moments a r e func- t i o n s o f which b i v a r i a t e gamma was used t h e r e s u l t s are q u i t e com- p a r a b l e u s i n g e i t h e r method.

    5. CONCLUSIONS AND SUMNARY

    Express ions f o r t h e p o s i t i v e i n t e g r a l moments a r e g i v e n f o r

    t h e r a t i o of c o r r e l a t e d gamma d i s t r i b u t e d v a r i a b l e s . Two under- l y i n g b i v a r i a t e gamma s t r u c t u r e s were cons ide red . It was shown

    t h a t t h e two s t r u c t u r e s p r o v i d e i d e n t i c a l r e s u l t s f o r a l l f i n i t e moments whenever t h e v a r i a b l e s a r e u n c o r r e l a t e d and a r e o n l y i d e n t i c a l f o r t h e f i r s t moment whenever t h e v a r i a b l e s a r e c o r r e l a t e d . A numer ica l e v a l u a t i o n f o r a s e l e c t e d s u b s e t o f t h e

    pa ramete r s p a c e i n d i c a t e d t h a t a l t h o u g h t h e h i g h e r moments do d i f f e r a c c o r d i n g t o t h e method used , t h e d i f f e r e n c e s a r e p robab ly i n s i g n i f i c a n t f o r most a p p l i c a t i o n s .

    Both methods r e v e a l t h e s i g n i f i c a n t e f f e c t t h a t c o r r e l a t i o n h a s upon t h e p o p u l a t i o n moments. Hence, b o t h methods r e i n f o r c e t h e need f o r e x a c t d i s t r i b u t i o n a l r e s u l t s f o r t h e r a t i o o f gamma d i s t r i b u t e d v a r i a b l e s a s i n d i c a t e d i n t h e s t u d y of r a i n f a l l o r r a i n s e e d i n g exper imen t s a s g i v e n i n F lueck and Hol land (1976).

    6 . BIBLIOGRAPHY

    Dwivedi, T . D. and Chaubey, Y . P . , (1981) . Moments of a r a t i o of two p o s i t i v e q u a d r a t i c forms i n normal v a r i a t e s . Commun. S t a t i s t . - S i m u l a . Computa., 10, 503-516.

    Dow

    nloa

    ded

    by [T

    he U

    nivers

    ity of

    Man

    ches

    ter L

    ibrary

    ] at 1

    3:58 2

    2 Octo

    ber 2

    013

  • MOMENTS FOR RATIO OF CORRELATED GAMMA VARIATES 259

    Gradshteyn, I. S. and Ryshik, I. M . , (1965). T a b l e s o f I n t e g r a l s , S e r i e s , and P r o d u c t s . New York: Academic P r e s s .

    F lueck , J . A. and Hol l and , B . S . , (1976) . R a t i o e s t i m a t o r s and some i n h e r e n t problems i n t h e i r u t i l i z a t i o n . J. Appl. Meteoy., 15, 535-543. -

    Flueck , J . A . , Ho l l and , B . S . , and Lee, R., (1975) . D i s t r i b u t i o n of t h e r a t i o of c o r r e l a t e d sums of gamma v a r i a b l e s . Proc. S o c i a l Sc ience S e c t i o n , ASA, Washington, D . C . , pp. 285-291.

    J e n s e n , D, R . , (1970) . The j o i n t d i s t r i b u t i o n of q u a d r a t i c forms and r e l a t e d d i s t r i b u t i o n s . A u s t r a l . J. of S t a t i s t . , 12, 13-22.

    Lee, R . , Ho l l and , B . S . , and F lueck , J. A . , (1979) . D i s t r i b u t i o n of a r a t i o of c o r r e l a t e d gamma random v a r i a b l e s . SIAM J. Appl. Math., 2, 304-320.

    d i e l k e , P. IJ. and F lueck , J. A , , (1976) . D i s t r i b u t i o n s of r a t i o s f o r some s e l e c t e d h i v a r i a t e probability f u n c t i o n s . 1976 Proc. S o c i a l S c i e n c e S e c t i o n , ASA, Washington, D.C. , pp. 608-613.

    Tubbs, J. D. and Smith , 0 . E . , (1984) . A n o t e on t h e r a t i o o f p o s i t i v e l y c o r r e l a t e d gamma v a r i a t e s . Commun. S t a t i s t . - T h e o r . Meth. , 14, 13-23.

    Received Januahq, 198 5; Revdined Septernbeh, 19ti5.

    Dow

    nloa

    ded

    by [T

    he U

    nivers

    ity of

    Man

    ches

    ter L

    ibrary

    ] at 1

    3:58 2

    2 Octo

    ber 2

    013